DeGrado, P. Leder, Science, 225, 687 (1984).
`8. H. Persson and P. Leder, in preparation.
`9. C. Prives, Y. Bech, H. Shore, J. Virol. 33, 689
`(1980).
`10. M. J. Hayman et al., Cell 32, 579 (1983); M. L.
`J. M. Bishop, J.
`Sealy,
`Privalsky,
`L.
`P.
`McGrath, A. D. Levinson, ibid., p. 1257.
`11. H. Land, L. F. Parada, R. A. Weinberg, Nature
`(London) 304, 5% (1980).
`12. L. T. Feldman and J. R. Nevins, Mol. Cell Biol.
`3, 829 (1983).
`13. T. R. Stewart, A. R. Belive, P. Leder, Science,
`in press.
`14. J. Battey et al., Cell 34, 779 (1983); W. W.
`Colby, E. Y. Chen, D. H. Smith, A. D. Levin-
`son, Nature (London) 301, 722 (1983); R. Watt
`et al., ibid. 303, 725 (1983); 0. Bernard, S. Cory,
`
`S. Gerondakis, E. Webb, J. M. Adams, EMBO
`J. 2, 2375 (1983).
`15. R. T. Sauer, R. R. Yocum, R. F. Doolittle, M.
`Lewis, C. 0. Pabo, Nature (London) 298, 447
`(1983).
`16. Reviewed in T. J. Kelly, Jr., Organization and
`Replication of Viral DNA, A. Kaplan, Ed. (CRC
`Press, Boca Raton, Fla., 1982), pp. 115-146.
`17. U. K. Laemmli, Nature (London) 227, 680
`(1970).
`18. H.P. acknowledges a fellowship from the Euro-
`pean Molecular Biology Organization. We are
`also grateful to E. I. du Pont de Nemours & Co.,
`Inc., and American Business for Cancer Re-
`search Foundation for their support.
`16 April 1984; accepted 26 June 1984
`
`A Cell Line Expressing Vesicular Stomatitis Virus
`Glycoprotein Fuses at Low pH
`Abstract. A stable cell line expressing a complementary DNA clone encoding the
`vesicular stomatitis virus glycoprotein fused andformed polykaryons at pH 5.5. The
`formation ofpolykaryons was dependent on the presence of glycoprotein anchored
`at the cell surface and could be prevented by incubation of cells with a monoclonal
`antibody to the glycoprotein. Fusion occurred at a pH 0.5 unit lower than that
`observed for cells infected with vesicular stomatitis virus.
`
`The VSV glycoprotein (G protein) is a
`single polypeptide chain that is held in
`the viral membrane by a domain of hy-
`drophobic amino acids near the carboxyl
`terminus (10). More than 95 percent of
`
`A virus particle must enter the host
`cell to grow. There are two ways in
`which enveloped viruses are known to
`enter the cell (1). Paramyxoviruses such
`as Sendai virus can enter through direct
`fusion (in a pH-independent manner) of
`the virion envelope with the plasma
`membrane of the cell (1, 2). The second
`path of entry, which is used by influenza
`virus (1, 3), Semliki Forest virus (SFV)
`(1, 3, 4), and vesicular stomatitis virus
`(VSV) (1, 3, 5, 6), is the internalization of
`virus particles in coated vesicles. The
`internalized vesicle is acidified, possibly
`after fusing with other intracellular vesi-
`cles (1, 3). The low pH in the vesicles
`containing virus particles causes fusion
`of the viral envelope with the membrane
`of that vesicle (3), and the viral nucleo-
`capsids are released into the cytoplasm.
`Direct evidence for a membrane fusion
`activity of viral glycoproteins has been
`obtained by expressing cloned comple-
`mentary DNA's (cDNA's) encoding the
`SFV glycoproteins El and E2 (7) and the
`influenza
`hemagglutinin
`glycoproteins
`HAI and HA2 (8). In each case the
`respective
`glycoproteins,
`ex-
`when
`pressed transiently in eukaryotic cells,
`were shown to promote cell-to-cell fu-
`sion at low pH. It is believed that the
`hydrophobic amino terminus of HA2 of
`influenza virus is required to promote
`membrane fusion (8). The E2 protein of
`SFV does not promote fusion alone, but
`when both El and E2 are present on the
`cell surface, fusion will occur at low pH
`(7, 9). It has been suggested that a hydro-
`phobic amino acid sequence near the
`amino terminus of El might play a cru-
`cial role in fusion (7, 9).
`17 AUGUST 1984
`
`each protein molecule is exposed on the
`surface of the virion. It has been ob-
`served that VSV-infected cells will fuse
`at low pH and that virus particles alone
`will promote cell-to-cell fusion at low pH
`(1). Cell fusion was thought to be mediat-
`ed by G protein at the cell surface.
`In the study reported here we attempt-
`ed to determine whether G protein, in
`the absence of other VSV proteins, will
`promote cell fusion at low pH. We previ-
`ously described a mouse cell line (CG1)
`that expresses VSV G protein at the cell
`surface (11). These cells are ideal for
`investigating the role of G protein in cell
`fusion because they express this protein
`in the absence of other viral proteins, for
`example the VSV matrix protein, which
`could affect fusion by interacting with G
`protein. The matrix protein may interact
`with the cytoplasmic domain of G pro-
`tein during virus maturation (12). Stable
`expression of G protein in CG1 cells has
`been established with a hybrid expres-
`sion vector that includes the SV40 early
`promoter (13), cDNA sequences encod-
`ing normal G protein (10), the SV40
`small t intron, SV40 early polyadenyla-
`tion signals (13), and the 69 percent
`subgenomic DNA transforming fragment
`
`Fig. 1. Formation of polykar-
`yons as a result of cell fusion
`at low pH. CGI or CTGI
`cells (1 x 106) were plated in
`50-mm tissue culture dishes.
`The cells were grown for 16
`hours in Dulbecco's modified
`Eagle's medium (DMEM) plus
`5 percent fetal bovine serum in
`an atmosphere containing 10
`percent CO2. The culture me-
`dium was removed and re-
`placed with 2 ml of prewarmed
`(37°C) fusion medium (1.85
`mM NaH2PO4 * H20,
`8.39
`mM NaHPO4 - 7H20, 2.5 mM
`NaCI), 10 mM Hepes, and 10
`mM 2-(N-mopholino)ethane-
`sulfonic acid). This buffer was
`adjusted to a final pH of 5.25.
`The cells were incubated in
`this for I minute; then the fu-
`sion medium was removed and
`replaced with DMEM plus 5
`percent fetal bovine serum. In
`this experiment the cultures
`were returned to an incubator
`(37°C and 10 percent C02) and
`examined for fusion 12 hours
`later as a convenience, but it
`should be noted that fusion
`could be detected within 2
`hours after incubation in fu-
`sion medium. (A) Polykaryon
`formation in CGI cells. (B)
`Typical field of CTGI cells,
`showing the absence of poly-
`karyon formation. The photo-
`graphs were taken with Polar-
`oid film (type 55) and an Olym-
`pus inverted microscope.
`721
`
`Downloaded from https://www.science.org at Yale University on October 15, 2023
`
`Page 1 of 4
`
`KELONIA EXHIBIT 1029
`
`

`

`Unlike the HA2 G protein of influen(cid:173)
`za, mature G protein has no obvious
`hydrophobic amino acid sequence at or
`near its amino terminus that could medi(cid:173)
`ate fusion (8, 16, 17). However, by anal(cid:173)
`ysis of the hydrophobic index of G pro- .
`tein with the method of Kyte and Doolit(cid:173)
`tle (18), several stretches of amino acids,
`in addition to the carboxyl
`terminal
`transmembrane domain and the amino
`terminal signal sequence, were located in
`G protein which have a marked hydro(cid:173)
`phobic character (Fig. 3). These stretch(cid:173)
`es of amino acids, between residues 120
`to 150, 190 to 210, and 300 to 360, could
`be involved in promoting membrane fu(cid:173)
`sion, possibly after low pH-induced con(cid:173)
`formational changes. Site-specific muta(cid:173)
`genesis in these regions may allow us to
`identify a specific domain involved in
`fusion.
`Why does the maximum fusion of CG 1
`cells happen at pH 5.5 while the fusion of
`the same cells infected with VSV occurs
`at pH 6.0; and why is the extent offusion
`limited to a small percentage of the total
`the entire
`population of cells and not
`monolayer? A difference has not been
`found between the pH at which cells
`expressing only influenza virus glyco(cid:173)
`proteins HAl and HA2 or SFV glycopro(cid:173)
`teins El and E2 fuse and the pH at which
`the respective virus-infected cells fuse
`(1). We have reported (11)
`that >95
`percent of CG I cells express levels of G
`protein on their cell surface that can be
`detected by immunofluorescence. How(cid:173)
`ever, the amount of this protein varied
`from cell to cell in the population. Analy(cid:173)
`sis of CG 1 cells by flow cytometry (11)
`indicated considerable variation in the
`amount of G protein present on the sur(cid:173)
`face of cells in the population. It seems
`likely that only the small fraction of cells
`with large amounts of G protein at the
`cell surface can initiate fusion. If this is
`true, then the extent of polykaryon for(cid:173)
`mation and the shift in the pH at which
`fusion occurs may simply reflect
`the
`difference in the amount of G protein at
`the cell surface of virus-infected cells as
`opposed to stably expressing CG1 cells.
`Alternatively, the ability of VSV-infect(cid:173)
`ed cells to fuse at pH 6 could reflect an
`interaction between G protein and anoth(cid:173)
`er VSV protein, for example the matrix
`protein, which then serves to alter the
`distribution of G protein at the cell sur(cid:173)
`face and thereby affect low pH-induced
`fusion. Investigators using temperature(cid:173)
`sensitive mutants of VSV have obtained
`contradictory results with respect to the
`involvement of VSV matrix protein in
`the formation of polykaryons at neutral
`pH (19,20).
`In conclusion, cells that express VSV
`
`SCIENCE. VOL. 225
`
`of bovine papilloma virus (14). A hybrid
`SV40-VSV G protein messenger RNA
`(mRNA) is transcribed from the SV40
`early promoter. The mRNA encodes au(cid:173)
`thentic G protein, which is processed,
`glycosylated, and transported normally
`to the cell surface (11).
`We incubated CGI cells in fusion me(cid:173)
`dium at pH 5.2 for I minute and then in
`normal medium for 2 to 12 hours. Poly(cid:173)
`karyons with as many as 30 nuclei were
`then observed (Fig. 1A), but were not
`seen in the nontransformed parental
`(CI27) cell line, which does not express
`G protein.
`To determine whether fusion requires
`G protein to be Hanchored" at the cell
`surface or can be induced by a secreted
`form of G protein, we looked for poly(cid:173)
`karyon formation with the transformed
`cell
`line CTG I. CTG I cells express a
`truncated (TG) form of G protein that
`lacks the transmembrane anchor se(cid:173)
`quence and is secreted (11). Cells that
`express TG protein did not fuse at low
`pH (Fig. IB). This implies that G protein
`must be anchored in the cell membrane
`to promote cell fusion and that low pH(cid:173)
`induced fusion is not a consequence of
`cell transformation by the vector. A lack
`of fusion activity for a secreted form of
`influenza HA protein has also been re(cid:173)
`ported (8).
`
`Since low pH causes fusion of VSV(cid:173)
`infected cells (1), we compared the low
`pH-induced fusion of VSV infected
`CGI, CTGI, and CI27 cells with that of
`uninfected CGI, CTGI, and CI27 cells.
`To do this we incubated VSV-infected
`and uninfected cells in fusion medium of
`pH 5.0 to 6.5. When uninfected CGI
`cells were examined, the pH at which
`maximum forntation of polykaryons oc(cid:173)
`curred was 5.5 (Fig. 2). However, after
`VSV infection, CGI, C127, and CTGI
`cells all showed maximum fusion at pH
`6.0 (Fig. 2). In addition, unlike virus(cid:173)
`infected CG I cells, in which the entire
`monolayer fused after incubation at low
`pH, polykaryon formation in uninfected
`CG 1 cell cultures was limited to defined
`regions of the monolayer.
`Treatment of CG I cells with a mono(cid:173)
`clonal antibody (13) against G protein
`(15), which neutralizes virus infectivity,
`prevented their fusion at low pH. This
`inhibition was specific for G protein be(cid:173)
`cause fusion was not prevented in paral(cid:173)
`lel experiments with a monoclonal anti(cid:173)
`body against the VSV matrix protein.
`This demonstrates that G protein plays a
`role in the fusion of CG 1 cells. Whether
`this represents an interaction with a do(cid:173)
`main on G protein involved in fusion or
`is simply a result of stenc hindrance is
`unknown.
`
`25
`
`!
`
`fI)
`
`~ 20
`
`~
`~ 15
`
`fI)
`~
`~ 10
`~
`8.
`a 5
`Q)
`~
`i
`
`~,.....-~---r----,..---....,..-___
`5
`5.5
`6.5
`6
`
`~
`J:
`
`40
`o
`_VSV G __
`/ \
`.c
`f\ ~
`~ protein
`0)( 20
`~ ~
`0 ~r---"..4-l-f~.......-~~~...-.J!i.."~_.-4-~IJfp.JtiJY''rt(~-+-.' fJ IaA~ja.
`~ ~ \It J
`if ~ v I
`~ .: -20
`rVV"'l" , ~V
`J
`-40
`
`---IL.-..
`
`...&..-_
`
`-------""""------........- -_ _--L.
`o
`100
`300
`200
`Sequence number
`Fig. 3. Relative hydrophobic index of G protein. The continuous average hydrophobicity along
`the length of VSV G protein (including the signal sequence) was determined as described by
`Kyte and Doolittle (18). Continuous average hydrophobicity was determined from the average
`hydrophobicity of a moving segment of seven amino acids, starting at the amino terminus and
`proceeding to the carboxyl terminus. Bars appear over the stretches of amino acids with a
`marked hydrophobic character.
`
`400
`
`500
`
`722
`
`Fig. 2. The pH-dependent fusion of VSV-
`infected and uninfected CG 1 cells. Twelve
`plates of CG 1 cells were prepared as de-
`scribed in the legend to Fig. 1, except that six
`plates were first incubated with VSV (Indiana
`serotype; multiplicity of infection, approxi-
`mately ten) for 30 minutes at 37°C. The virus
`inoculum was removed and replaced with
`DMEM plus 5 percent fetal bovine serum.
`Sixteen hours later the cells were incubated
`with fusion medium at pH 4.7 to 6.5. Six
`parallel plates of CG 1 cells were treated simi-
`larly except that they were not incubated with
`virus. Fusion of uninfected CGl cells and
`virus-infected cells was determined 2 hours
`after incubation in fusion medium by counting
`the number of polykaryons with more than
`pH
`'four nuclei in 20 random fields of approximately 350 cells each. Fusion of VSV-infected cells is
`shown as the percentage of fused cells visible.
`
`"CD
`
`fI)
`
`100 ~
`.!
`Q)
`()
`
`U.
`
`CD
`
`75
`
`50
`
`"CD
`!c:::
`"j
`fI)
`2
`-;
`25 a
`atas
`'E
`CD
`()
`Q)
`CL
`
`Downloaded from https://www.science.org at Yale University on October 15, 2023
`
`Page 2 of 4
`
`

`

`G protein on their surfaces can, in the
`absence of other virus proteins, be used
`to investigate the ability of VSV G pro(cid:173)
`tein to cause membrane fusion. Fusion
`low pH and in the
`was observed at
`absence of any other virus-specific pro(cid:173)
`teins. The formation of polykaryons re(cid:173)
`quired that G protein be anchored at the
`cell surface and was specifically inhibit(cid:173)
`ed by monoclonal antibodies to G pro(cid:173)
`tein.
`
`R. Z. FLORKIEWICZ
`J. K. ROSE
`Molecular Biology and Virology
`Laboratory, Salk Institute,
`San Diego, California 92138
`
`References and Notes
`1. J. White, M. Kielian, A. Helenius, Q. Rev. Bioi.
`Phys. 16 (No.2), 151 (1983).
`2. M. Homma and M. Ohuchi, J. Virol. 12, 1457
`(1973).
`3. M. Marsh et al., Cold Spring Harbor Symp.
`Quant. Bioi. 156, 835 (1982).
`4. A. Helenius, J. Kartenbeck, K. Simons, E.
`Fries, J. Cell Bioi. 84, 404 (1980).
`5. J. E. Dahlberg, Virology 58, 250 (1974).
`6. D. P. Fan and B. M. Sefton, Cell IS, 985 (1978).
`7. C. Kondor-Koch, B. Burke, H. Garotf, J. Cell
`Bioi. 97, 644 (1983).
`
`8. J. White, A. Helenius, M.-J. Gething, Nature
`(London) 300, 658 (1982).
`9. H. Garotf, A.-M. Frischauf, K. Simons, H.
`Lehrach, H. Delius, ibid. 288, 236 (1980).
`10. J. K. Rose and J. E. Bergmann, Cell 30, 753
`(1982).
`11. R. Z. Florkiewicz, A. Smith, J. Bergmann, J. K.
`Rose, J. Cell BioI. 97., 1381 (1983).
`12. H. Fraenkel-Conrat and R. R. Wagner, Compre(cid:173)
`hensive Virology (Plenum, New York, 1975),
`pp. 1-93.
`13. R. C. Mulligan and P. Berg, Science 209, 1422
`(1980).
`14. D. R. Lowy et al., Nature (London) 287, 72
`(1980).
`.
`15. L. Lefrancois and D. Lyles, Virology 121, 157
`(1982).
`16. J. J. Skehel and M. D. Waterfield, Proc. Natl.
`Acad. Sci. U.S.A. 72, 93 (1975).
`17. J. J. Skehel, P. M. Bayley, E. B. Brown, S. R.
`Martin, M. D. Waterfield, J. M. White, I. A.
`Wilson, D. C. Wiley, ibid. 79, 968 (1982).
`18. J. Kyte and R. F. Doolittle, J. Mol. Bioi. 157,
`105 (1982).
`19. F. Chany-Foumier, C. Chany, F. Lafay, J. Gen.
`Virol. 34, 305 (1977).
`20. K. Handa, F. Chany-Foumier, S. Rousset, C.
`Chany. BioI. Cell 44, 261 (1982).
`21. We thank J. White for providing advice on the
`protocol for fusion: D. Lyles for monoclonal
`antibodies: R. Doolittle for performing the com(cid:173)
`puter analysis of the VSV G protein; and L.
`Zokas, C. Machamer, B. Sefton, and T. Hunter
`for helpful suggestions concerning the manu(cid:173)
`script. Supported by grants from the Public
`Health Service (AI1548l) and the National Can(cid:173)
`cer Institute (CA 14195) and by a Public Health
`Service fellowship (1 F32 AI06911-0l) to R.Z.F.
`
`17 February 1984; accepted 6 June 1984
`
`Immunologically Induced Alterations of Airway
`Smooth Muscle Cell Membrane
`
`Abstract. Active and passive sensitization, both in vivo and in vitro, caused
`significant hyperpolarization of airway smooth muscle cell preparations isolated
`from guinea pigs. An increase in the contribution of the electrogenic Na+ pump to
`the resting membrane potential was responsible for this change. Hyperpolarization,
`as induced by passive sensitization, was not prevented by agents that inhibit specific
`mediators ofanaphylaxis but was abolished when serumfrom sensitized animals was
`heated. The heat-sensitive serumfactor, presumably reaginic antibodies, appears to
`be responsible for the membrane hyperpolarization of airway smooth muscle cells
`after sensitization.
`
`A number of respiratory diseases, in(cid:173)
`cluding bronchial asthma, are character(cid:173)
`ized by an increased bronchoconstrictive
`response to numerous stimuli such as
`histamine or methacholine inhalation.
`
`The physiological factors underlying this
`so-called airway hyperreactivity are
`poorly understood. One idea is that fun(cid:173)
`damental changes may occur in the ex(cid:173)
`citability and contractile properties of
`
`the airway smooth muscle itself (1). We
`showed that, in guinea pigs, sensitization
`with ovalbumin is associated with hyper(cid:173)
`polarization of airway smooth muscle
`and that
`this hyperpolarization is,
`in
`turn, related to an increase in the contri(cid:173)
`bution of the electrogenic Na+ pump to
`the resting membrane potential. Further,
`hyperpolarization of the airway smooth
`muscle could be produced by passive
`sensitization in vitro and was not pre(cid:173)
`vented by agents that inhibit mediators
`of anaphylaxis. However, heating serum
`obtained from sensitized animals pre(cid:173)
`vented the change in resting membrane
`potential. These findings suggest that the
`airway response induced by sensitization
`to antigen involves a direct interaction
`between specific serum antibodies and
`the airway smooth muscle cell mem(cid:173)
`brane. Thus, in addition to the role of the
`vagal reflex (2),
`release of mediators
`from mast cells (3), and possible alter(cid:173)
`ation of specific membrane receptors (4),
`changes in airway smooth muscle mem(cid:173)
`brane can be responsible for the phe(cid:173)
`nomenon of airway hyperreactivity.
`Segments of the middle portion of tra(cid:173)
`chea isolated from male guinea pigs of
`the Camm-Hartley strain were studied in
`a temperature-controlled chamber as de(cid:173)
`scribed (5). Single smooth muscle cells
`of tracheal muscle were impaled with
`glass microelectrodes made of borosili(cid:173)
`cate glass filled with 3M KCI and having
`less than 10mV and
`a tip potential
`resistance of 80 to 90 megaohms. The tip
`potential and the resistance of each elec(cid:173)
`trode were measured after each impale(cid:173)
`ment. Successful impalement of a cell
`was indicated by a prompt negative de(cid:173)
`flection of the oscilloscope trace and
`maintenance of a steady potential (within
`5 mY) for at least 10 seconds (6). Simul(cid:173)
`taneously with the measurement of rest(cid:173)
`ing membrane potential (Em)' the isomet(cid:173)
`ric force developed by tracheal segments
`was measured by means of a special
`
`Table 1. The effect of active sensitization, active sensitization and resensitization, passive in vivo and in vitro sensitization on the resting
`membrane potential of guinea pig airway smooth muscle cells, and the response of airway smooth muscle preparations to ovalbumin, ouabain
`(IO-sM) and K+·free solution. Values are means ± standard error; N.R., no response; N.D., not done.
`
`Condition
`
`Em (mV)
`
`Controls
`Active sensitization
`Active sensitization and re-
`sensitization
`Controls
`Passive sensitization in vivo
`Controls
`Passive sensitization in vitro
`*p < 0.001 compared to control.
`
`17 AUGUST 1984
`
`-61.3 ± 0.5
`-72.7 ± 0.6*
`-78.1 ± 0.5*
`
`-60.5 ± 0.4
`-69.5 ± 0.3*
`-60.7 ± 0.6
`-68.5 ± 0.4*
`
`Peak response to ovalbumin
`
`Em (mV)
`
`Em
`(mV)
`
`N.R.
`-S6.3 ± 0.3
`-53.7 ± 0.8
`
`N.R.
`-51.2 ± 1.3
`N.R.
`-53.0 ± 0.9
`
`Peak
`isometric
`force (g)
`N.R.
`3.8 ± 0.3
`7.8 ± 0.4
`
`N.R.
`S.3 ± 1.1
`N.R.
`4.7 ± 0.7
`
`After
`10-5M
`ouabain
`
`-49.3 ± 0.6
`-51.8 ± 0.5
`-49.9 ± 0.9
`
`N.D.
`-50.5 ± 0.8
`-49.5 ± 5
`-50.2 ± 0.6
`
`After
`K+·free
`solution
`
`-50.7 ± 0.4
`-51.9 ± 0.6
`-51.4 ± 0.5
`
`N.D.
`N.D.
`N.D.
`N.D.
`
`723
`
`Downloaded from https://www.science.org at Yale University on October 15, 2023
`
`Page 3 of 4
`
`

`

`A Cell Line Expressing Vesicular Stomatitis Virus Glycoprotein Fuses at Low
`pH
`R.Z. Florkiewicz and J. K. Rose
`
`Science 225 (4663), . DOI: 10.1126/science.6087454
`
`View the article online
`https://www.science.org/doi/10.1126/science.6087454
`Permissions
`https://www.science.org/help/reprints-and-permissions
`
`Use of this article is subject to the Terms of service
`
`Science (ISSN 1095-9203) is published by the American Association for the Advancement of Science. 1200 New York Avenue NW,
`Washington, DC 20005. The title Science is a registered trademark of AAAS.
`
`Downloaded from https://www.science.org at Yale University on October 15, 2023
`
`Page 4 of 4
`
`